On 28.05.20 21:59, Julien Pivotto wrote: > > Should we recommend kelvin instead of celcius in histograms and > summaries as there would not be negative observations ? > https://github.com/prometheus/prometheus/issues/6669
Technically correct answer: We should just fix histograms. Nerdy answer: https://en.wikipedia.org/wiki/Negative_temperature Pragmatic answer: Temperature histograms almost never happen in practice. That's why I avoided that example when arguing why negative observations are a thing. A normal temperature measurement will happen as a gauge. When would you sample temperature observations in a histogram? I think that only happens if you have other events, which you then want to correlate with temperature. (Made-up example from the online-serving world: You have created a mobile app for users to check in at places. It also sends out the temperature with the check-in requests for some reason. Then you could create a histogram of check-in temperatures.) -- Björn Rabenstein [PGP-ID] 0x851C3DA17D748D03 [email] [email protected] -- You received this message because you are subscribed to the Google Groups "Prometheus Developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-developers/20200529123806.GP2326%40jahnn.
